Papers with Code 아카이브

Safety Alignment

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

전체 기법전체 데이터셋

하위 태스크

0

평가표

0

보존된 평가 기록

0

코드

관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.

https://github.com/git-disl/safety-tax

공식 코드

Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable

https://github.com/git-disl/virus

공식 코드

Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation

https://github.com/guanzihan/benign-samples-matter

공식 코드

Benign Samples Matter! Fine-tuning On Outlier Benign Samples Severely Breaks Safety

https://github.com/haneul-yoo/csrt

공식 코드

Code-Switching Red-Teaming: LLM Evaluation for Safety and Multilingual Understanding

https://github.com/hanjianghu/nbf-llm

공식 코드

Steering Dialogue Dynamics for Robustness against Multi-turn Jailbreaking Attacks

https://github.com/horizonsinzqs/queryattack

공식 코드

QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language

https://github.com/humainlab/understand_marginpo

공식 코드

A Common Pitfall of Margin-based Language Model Alignment: Gradient Entanglement

https://github.com/hwanchang00/CoPriva

공식 코드

Keep Security! Benchmarking Security Policy Preservation in Large Language Model Contexts Against Indirect Attacks in Question Answering

https://github.com/jarviswang94/multilingual_safety_benchmark

공식 코드

All Languages Matter: On the Multilingual Safety of Large Language Models

https://github.com/jc-ryan/agentalign

공식 코드

AgentAlign: Navigating Safety Alignment in the Shift from Informative to Agentic Large Language Models

https://github.com/jiahuigeng/vscbench

공식 코드

VSCBench: Bridging the Gap in Vision-Language Model Safety Calibration

https://github.com/justincui03/or-bench

공식 코드

OR-Bench: An Over-Refusal Benchmark for Large Language Models

https://github.com/juzhengz/lori

공식 코드

LoRI: Reducing Cross-Task Interference in Multi-Task Low-Rank Adaptation

https://github.com/kalyan-nakka/bitbypass

공식 코드

BitBypass: A New Direction in Jailbreaking Aligned Large Language Models with Bitstream Camouflage

https://github.com/kizna1ver/dream

공식 코드

DREAM: Disentangling Risks to Enhance Safety Alignment in Multimodal Large Language Models

https://github.com/lenijwp/eclipse

공식 코드

Unlocking Adversarial Suffix Optimization Without Affirmative Phrases: Efficient Black-box Jailbreaking via LLM as Optimizer

https://github.com/llbao/arrattack

공식 코드

One Model Transfer to All: On Robust Jailbreak Prompts Generation against LLMs

https://github.com/llm-tuning-safety/llms-finetuning-safety

공식 코드

Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

https://github.com/lslland/t-vaccine

공식 코드

Targeted Vaccine: Safety Alignment for Large Language Models against Harmful Fine-Tuning via Layer-wise Perturbation

https://github.com/lsvih/attackPrefixTree

공식 코드

Exploiting Prefix-Tree in Structured Output Interfaces for Enhancing Jailbreak Attacking

https://github.com/m3rg-iitd/aila

공식 코드

Autonomous Microscopy Experiments through Large Language Model Agents

https://github.com/mansicer/Q-Adapter

공식 코드

Q-Adapter: Customizing Pre-trained LLMs to New Preferences with Forgetting Mitigation

https://github.com/maxwe11y/gaslightingllm

공식 코드

Can a large language model be a gaslighter?

https://github.com/mickelliu/selfplay-redteaming

공식 코드

Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models

https://github.com/microsoft/test_awareness_steering

공식 코드

Linear Control of Test Awareness Reveals Differential Compliance in Reasoning Models

https://github.com/neuralsentinel/safeinfer

공식 코드

SafeInfer: Context Adaptive Decoding Time Safety Alignment for Large Language Models

https://github.com/neuralsentinel/soteria

공식 코드

Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment

https://github.com/ny1024/race

공식 코드

Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models

https://github.com/openfeedback/superhf

공식 코드

SuperHF: Supervised Iterative Learning from Human Feedback

https://github.com/openstellarteam/bsa

공식 코드

Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Models

https://github.com/osu-nlp-group/amplegcg

공식 코드

AmpleGCG: Learning a Universal and Transferable Generative Model of Adversarial Suffixes for Jailbreaking Both Open and Closed LLMs

https://github.com/osu-nlp-group/eval-llm-trust

공식 코드

How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities

https://github.com/pandragonxiii/cider

공식 코드

Cross-modality Information Check for Detecting Jailbreaking in Multimodal Large Language Models

https://github.com/pikepokenew/IRR

공식 코드

Separate the Wheat from the Chaff: A Post-Hoc Approach to Safety Re-Alignment for Fine-Tuned Language Models

https://github.com/pillowsofwind/course-correction

공식 코드

Course-Correction: Safety Alignment Using Synthetic Preferences

https://github.com/pipilurj/mllm-protector

공식 코드

MLLM-Protector: Ensuring MLLM's Safety without Hurting Performance

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.